HLA-G MOLECULES: FROM EMBRYO IMPLANTATION TO OOCYTE MATURATION

Università degli Studi di

Ferrara

DOTTORATO DI RICERCA IN

Biochimica, Biologia Molecolare e Biotecnologie

CICLO

XXI

COORDINATORE Prof. BERNARDI FRANCESCO

HLA-G MOLECULES:

FROM EMBRYO IMPLANTATION

TO OOCYTE MATURATION

Settore Scientifico Disciplinare MED/03

Dottorando Tutore

Dott. Rizzo Roberta Prof. Baricordi Olavio

It is not because things are difficult

that we do not dare,

it is because we do not dare

that they are difficult.

The present thesis is based on the following publications:

I. This thesis is submitted to Expert Review Obstet. Gynecol. (2009).

II. Rizzo R, Baricordi OR. HLA-G expression and regulation in early embryos. Am. J. Reprod. Immunol. 56 (1): 17 (2006).

III. Hviid TVF, Rizzo R, Melchiorri L, Stignani M, Baricordi OR. Polymorphism in the 5’-upstream regulatory and 3’-untranslated regions of the HLA-G gene in relation to soluble HLA-G and IL-10 expression. Human Immunol. 67 (1-2), 53-62 (2006).

IV. Rizzo R, Melchiorri L, Stignani M, Baricordi OR. HLA-G expression is a fundamental prerequisite to pregnancy. Human Immunol. 68 (4), 244-250 (2007).

V. Rizzo R, Fuzzi B, Stignani M, Criscuoli L, Melchiorri L, Dabizzi S, Campioni D, Lanza F, Marzola A, Branconi F, Noci I, Baricordi OR. Soluble HLA-G molecules in follicular fluid: a tool for oocyte selection in IVF? J. Reprod. Immunol. 74 (1-2), 133-142 (2007).

VI. Baricordi OR, Stignani M, Melchiorri L, Rizzo R. HLA-G and inflammatory diseases. Inflamm. Allergy Drug Targets. 7 (2), 67-74 (2008).

VII. Borgatti M*, Rizzo R*, Canto MB, Fumagalli D, Renzini MM, Fadini R, Stignani M, Baricordi OR, Gambari R. Release of sICAM-1 in oocytes and in vitro fertilized human embryos. PLoS ONE 3 (12), e3970 (2008). *The first two authors contributed equally to the work.

VIII. Rizzo R, Stignani M, Amoudruz P, Nilsson C, Melchiorri L, Baricordi O, Sverremark Ekstrom E. Allergic women have reduced sHLA-G plasma levels at delivery. Sumbmitted to Am. J. Immunol. (2008).

IX. Rizzo R, Dal Canto MB, Stignani M, Fadini R, Fumagalli D, Mignini Renzini M, Borgatti M, Gambari R, Baricordi OR. Production of sHLA-G molecules by “in vitro” matured cumulus-oocyte complex. Sumbmitted to

Preface

The work included in this thesis was performed during my Ph.D. studies at the Department of Experimental and Diagnostic Medicine – Section of Medical Genetics.

First of all I am grateful to Professor Olavio Baricordi who have introduced me to the research and offered me many opportunities. Warm thanks to my labmates with whom I have been sharing everyday life and all the excitements and disappointments of the research. I heartly thank Lory for her advices and sincere friendship; Marina for constructive talks, not only about work, and for keeping me in touch with molecular biology; Teresa for her resourceful helpfulness; Alessandra for her support. Beyond strictly scientific matters, I am heartly grateful to all these people for their friendship.

This thesis would never have been possible without the constant support of a number of collaborators. A very special thank is given to Dr. Thomas Hviid, from the Department of Clinical Biochemistry of the Roskilde University with whom I have had many good discussions; to Professor Gambari and Dr. Monica Borgatti from the Biochemistry and Molecular Biology of the University of Ferrara and to Dr. Lanza and Dr. Diana Campioni from the Hematology of the University of Ferrara, for the idea exchanges and technical support.

Finally, and most of all, I want to thank my family, Mum, Dad and Stefano, for their support and understanding. I want to share with them the joy of this event of my life.

Contents

Pages

HLA-G molecules in ART: from embryo implantation

to oocyte maturation

Introduction

2. Human Major Histocompatibility Complex ..………. 3

3. The non classical HLA class Ib genes ……….. 4

4. The HLA-G antigen ……… 4

4.1. The HLA-G gene ………. 4

4.2. HLA-G expression and function ………... 8

5. HLA-G and pregnancy ……….. 12

5.1. HLA-G and trophoblasts ……….. 15

5.2. HLA-G and embryo ………... 18

5.3. HLA-G and oocyte ………... 27

Conclusions and perspectives

7.

Acknowledgments

References

Abstract

Riassunto (Abstract in Italian)

HLA-G molecules in ART: from embryo implantation to oocyte

maturation

Roberta Rizzo†

Department of Experimental and Diagnostic Medicine – Section of Medical Genetics – University of Ferrara, Ferrara (Italy)

Running title: HLA-G molecules in ART

†Author for correspondence

Department of Experimental and Diagnostic Medicine – Section of Medical Genetics – University of Ferrara – Via Luigi Borsari 46 – 44100 Ferrara – Italy

Tel ++390532455383 Fax ++390532455380 e-mail rbr@unife.it

Abstract

Pregnancy is commonly considered a semi-allograft as half of the fetal genome derives from the father. However in normal pregnancy several tolerance mechanisms have been demonstrated to counteract the maternal immune response. Among these, the expression of HLA-G by invasive cytotrophoblasts has shown to play a fundamental role in creating a tolerogenic condition at the feto-maternal interface. The possible role of soluble HLA-G molecules as a marker for oocyte/embryo selection is reviewed comparing the contrasting results present in the literature and the significance of HLA-G modulation in assisted reproduction.

Key words: assisted reproduction, HLA-G, embryo, oocyte, marker

1. Actual problems of ART

The present efficiency of assisted reproduction (ART) seems to be far from desired. Until now the probability of a successful pregnancy following an in vitro fertilization procedure (IVF) is approximately 18%, with a baby rate of at about 14%. Since the birth of the first baby created by ART, Louise Brown in England in 1978 [1], it has produced more than 40,000 ART babies each year. However the average success rate is still low because of the inability to assess embryo quality using the currently available biochemical, genetic and imaging methods. At the present time it is not possible to determine for sure the best oocyte to fertilize or the most appropriate embryo to transfer. Therefore the usual practice is to transfer two or three embryos to improve the chance of a positive pregnancy outcome. Women undergoing ART have a 20-fold increased risk of twins and a 400-fold increased risk of triplets or higher order pregnancies with a perinatal mortality and morbidity [2,3]. Several studies are involved in the identification of non-invasive methods to determine the

improving pregnancy rates. Since the beginning of this new century the value of non classical HLA (Human leukocyte antigen) class I – G antigen as a marker of oocyte/embryo competency is under debate.

2. Human Major Histocompatibility Complex

The human Major Histocompatibility Complex (MHC) is a set of molecules encoded by a series of genes (~130) located on the short arm of chromosome 6 that are responsible for lymphocyte recognition, "antigen presentation" and immune response regulation. This gene complex comprises several distinct loci, grouped closely together on a 4-6 Mb chromosomal segment. In humans they are called HLA for Human Leucocyte Antigens. These antigens can be subdivided into three major classes: class I, class II and class III (Figure 1). The class I and class II antigens are expressed on cells and tissues whereas class III antigens are mainly serum and body fluid proteins (e.g.C4, C2, factor B, TNF, complement components). The class I gene complex contains three major loci A, B and C. Each of these loci encodes for an alpha-chain polypeptide that associates to ß2-microglobulin, encoded by a gene on chromosome 15. The class II gene complex contains at least three loci, DP, DQ and DR; each of these loci encodes for one alpha- and one beta-chain polypeptide which associate together to form the class II antigens. Since the end of the 80’s a new group of antigens has become interesting: the non classical HLA class I molecules [4,5].

Figure 1. Schematic representation of the HLA region on chromosome 6. The

DP DQ DR B C E A G F

Class II Class III Class I

3. The non classical HLA class Ib genes

Non-classical MHC class Ib molecules are closely homologous to classical class Ia molecules but are distinguished by their limited polymorphism and low cell surface expression. The class Ib molecules are not just vestigial evolutionary remnants of classical class Ia molecules; rather some are involved in highly specialized roles, as testified by their conservation between different species. The duo comprised of HLA-E in human, Qa-1 in mouse and HLA-G in human, Qa-2 in mouse constitutes a clear homology between species. In 1993, Warner et al. [6] demonstrated a reproductive advantage in mice encoding Qa-2 molecules by preimplantation embryo development (Ped) gene, the homologue of human HLA-G gene. The Qa-2 antigen has been detected on murine oocytes and early-cleavage and blastocyst-stage embryos where it seems to function as a mediator of mitogenic signals between embryo and uterine environment [7]. For these characteristics, Qa-2 and HLA-G antigens seem to share not only structural but also functional similarities in the regulation of immune response, through interaction with both inhibitory and activatory receptors [8,9].

4. The HLA-G antigen

HLA-G antigen is a non-classical HLA class I molecule characterized by (i) a low allelic polymorphism, (ii) a restricted tissue distribution to trophoblasts and a subset of thymic epithelial cells, (iii) mRNA alternative splicing that generates seven proteic isoforms and (iv) a tolerogenic and anti-inflammatory biological function [10].

4.1. The HLA-G gene

The HLA-G gene features low allelic polymorphisms with 36 HLA-G alleles acknowledged in the coding region (http//:www.anthonynolan.org.uk/HIG/). HLA-G gene is also polymorphic at the 5’-upstream regulatory region (5’ URR) and at the

expression [11]. A 14 bp insertion/deletion polymorphism (rs16375) in exon 8 in the 3’ UTR has been reported and associated with mRNA stability and HLA-G protein expression [12,13] (Figure 2). The allele with an insertion of 14 bp has been associated with lower levels of HLA-G expression than the allele with the 14 bp deleted [13-15]. An additional alternatively spliced HLA-G transcripts lacking 92 bp of the first part of exon 8 is observed within the insertion of 14 bp allele and is characterized by a more stable transcript [16].

Transactivation of classical MHC class I genes is mediated by two groups of juxtaposed cis -acting regulatory modules: (i) the up-stream enhancer A and ISRE (interferon-sensitive response element) which mediate the constitutive and cytokine-induced expression; (ii) the S-X-Y module which controls the constitutive and CIITA (class II transactivator) mediated transactivation. These modules are divergent in HLA-G gene that is unresponsive to NF-kappaB (nuclear factor-kappaB), IRF-1 (interferon regulatory factor 1), and CIITA mediated induction pathways [17]. The HLA-G gene promoter shows a putative interferon-regulatory factor (IRF)-1 binding site 746 base pairs upstream from ATG, which is distinct from the interferon-responsive element within proximal class Ia gene promoters. This control region is the putative element which mediates interferon beta-induced expression of the HLA-G gene [18]. The HLA-HLA-G promoter contains three cAMP/PMA response elements (CRE/TRE) with binding affinity for REB (rice endosperm bZIP)/ATF (activating transcription factor-2) and Fos/Jun proteins. It has been shown that HLA-G transactivation is regulated by CREB (cAMP-response element-binding protein), CREB-binding protein (CBP), and p300. These features represent the unique regulation of HLA-G transcription among the MHC class I genes [19].

Epigenetic mechanisms seem to play an important role in HLA-G expression [20,21]. The potential role of DNA methylation on HLA-G expression has been tested in human tumours considering the effect of the methylation inhibitor deoxycytidine on the CpG-enriched regulatory region of the HLA-G gene. The 5-aza-dC treatment results in hypomethylation of putative control sequences within the 5'

regulatory region of HLA-G and these changes in methylation correlate with a significant increase in expression.

The HLA-G gene seems not to undergo genomic imprinting, in fact it is co-dominantly expressed on trophoblast cells [22].

A post-transcriptional regulation for HLA-G molecules is also possible because of the expression in advance of the molecules that are essential for cell surface expression of class I molecules, ß2-microglobulin (ß2m) and the transporter for

antigen processing proteins (TAP1 and TAP2), allowing a rapid accumulation of HLA-G protein in differentiating extravillous cytotrophoblast cells [23].

Figure 2. Multiple HLA-G proteins derived from alternative splicing of HLA-G mRNA.

Upper: The HLA-G gene is composed of 8 exons. The α, intracellular and transmembrane domains and the 14 bp insertion/deletion polymorphism (rs16375) in exon 8 in the 3’ untranslated region (UTR) are represented. The gene is alternatively spliced to yield 7 transcripts. In two of these, a stop sequence in intron 4 results in soluble isoforms.

Lower: The 7 HLAG proteic isoforms: four membranebound (HLAG1, G2, G3, -G4) and three soluble (HLA-G5, -G6, -G7) molecules.

4.2. HLA-G expression and function

In physiological conditions HLA-G protein presence is restricted to trophoblasts, thymus, cornea, nail matrix, pancreas, erythroid and endothelial precursors [24]. Unlike HLA class Ia antigens, seven HLA-G isoforms are generated by alternative splicing of its primary transcript. Four of them, HLAG1,-G2, -G3 and -G4, are membrane-bound, while three, HLA-G5, -G6 and -G7, are soluble molecules (Figure 2). The soluble isoforms retain the intron 4 which includes a stop codon and leads to the termination of the mRNA translation before the transmembrane domain. The HLA-G1 and HLA-G5 structures are characterized by three alpha domains, differently from the other isoforms which lack one or more globular domain. The most analyzed isoforms are HLA-G1 and HLA-G5 antigens. The proteolytical cleavage of surface isoform HLA-G1 generates the soluble HLA-G1 form (sHLAG1) [25].

An in frame termination codon in HLA-G exon 6 leads to a truncated cytoplasmic tail which is 19 amino acids shorter than the corresponding tails of HLA-A, -B and –C proteins. This feature prevents the signal transduction from the cell surface to the nucleus. However, the membrane-bound HLA-G can localize in lipid rafts and can act as a signaling molecule, via modification of the phosphorylation state of raft-localized proteins [26].

The HLA-G production up-regulation is controlled by interleukin (interleukin-10), interferon and hormone molecules [13].

Membrane-bound HLA-G1 and soluble HLA-G (HLA-G5 and sHLA-G1) molecules exert immunosuppressive effects: (i) inhibit the cytotoxic activity of CD8 positive T lymphocytes (CTL) and Natural Killer (NK) cells [27], (ii) induce the apoptosis of NK and activated cytotoxic T cells [28], (iii) inhibit the allogeneic CD4 positive T-cell proliferation and interfere with naïve CD4 T-T-cell priming [29], (iv) inhibit antigen presenting cell and B lymphocyte differentiation [30], (v) induce regulatory T cells [31] (Figure 3). Furthermore, sHLA-G affects angiogenesis interacting with

endothelial cells [32] and induces resting NK cells to produce chemokines and cytokines [8].

The functions of HLA-G molecules are due to their ability to act as a ligand for different receptors expressed by immune cells (Figure 3). HLA-G interacts with NK receptor KIR2DL4 [8] and leukocyte inhibitory receptors (LILRs) / immunoglobulin-like transcripts (ILT) [9] as LILRB1 (LIR-1/ILT2/CD85j), which is highly expressed on T and B lymphocytes and with LILRB2 (LIR-2/ILT4/CD85d), present mainly in monocytes/macrophages. The alpha3 domain of HLA-G is the putative binding site for ILT receptors [33] while the residues Met76 and Gln79 in the alpha1 domain play a critical role in the recognition of KIR2DL4 receptor [34].

Soluble HLA-G has potentially a higher range of activity than membrane-bound HLA-G. The circulating isoforms could bind to the same sets of leukocytes and perform exactly the same functions also systemically.

The membrane-bound and soluble HLA-G proteins have monomer, dimer, and oligomer forms; the dimer seems to have a dominant effect on the LILRB signaling. Dimers of HLA-G have been observed on the surface of transfected cells [35], on choriocarcinoma cell line JEG-3 [36] and on first trimester trophoblast cells [37]. A disulfide-bonded dimer conformation is possible for the presence of a cysteine 42 residue that is present only in the heavy chain α1 domain of HLA-G [33,35].

Soluble HLA-G1 is able to inhibit endothelial cells through specific interaction with the CD160 molecule, a glycosylphosphatidylinositol-anchored, major histocompatibility complex (MHC) Class I-dependent, immunoglobulin-like receptor, that is expressed by activated endothelial cells [32] (Figure 3). This interaction seems to lead to apoptosis of endothelial cells required for normal placental development.

HLA-G molecules undergo the trogocytosis mechanism: some effectors CD4 and CD8 T lymphocytes acquire immunosuppressive HLA-G1 molecules from antigen presenting cell membranes and reverse their function from effectors to regulatory cells [38].

The HLA-G expression has been analyzed in different pathological conditions, such as transplantation, oncology, viral infections, inflammatory and autoimmune diseases indicating that HLA-G can favour graft tolerance, tumor cell and virally infected immune escape and control the inflammatory conditions [39].

Figure 3. HLA-G receptors. HLA-G receptors expressed on immune (CD8 T and CD4 T cells, B cells, natural killer (NK) cell, macropages, dendritic cells) and endothelial cells. ILT: immunoglobulin-like transcript; KIR: killer inhibitory receptor; TCR: T cell receptor.

α2 α 3 α1 β2m α2 α 3 α1 β2m

CD8+ T lymphocyte

CD4+ T lymphocyte

B lymphocyte

Macrophage

TGF-ß1

Dendritic cell

Endothelial cell

Regulatory T cell

NK cell

5. HLA-G and pregnancy

Pregnancy is commonly considered a paradox. In an outbreed population, half of the fetal genes are paternal, thus the fetus may be considered a semi-allograft. However in normal pregnancy the maternal immune system does not reject the fetus, even if it is a “stranger in a strange land” [40], as a series of tolerogenic mechanisms are developed to allow gestation and birth of healthy babies.

The enigma of the absence of fetal rejection was stated in 1953 by Sir Peter Medawar. He proposed the presence of protective mechanisms that could stimulate the tolerance in the mother [41]. Especially, the absence of HLA class I and II molecules modulation and the production of immuno-suppressant soluble molecules (progesterone, prostaglandine, transforming growth factor–B1, Interleukin-10) [42] by the fetal tissues strongly confirm the development at the feto-maternal interface of tolerogenic conditions.

Recently a fundamental role in maintaining this tolerogenic condition has been proposed for HLA-G antigens. The soluble HLA-G molecules have been detected in the plasma of pregnant women with increased levels during the first trimester in comparison to non-pregnant women [43] (Table 1). On the contrary sHLA-G plasma levels decrease during the third trimester [44,45] while it has an impressive boost at delivery [46] probably deriving from the shedding of placental membrane-bound HLA-G molecules.

sHLA-G concentrations in serum/plasma of pregnant women have been associated with clinical outcome. sHLA-G levels in plasma from women who subsequently develop preeclampsia, a potentially dangerous disorder of human pregnancy associated with utero-placental vascular defects [47], and/or intrauterine growth retardation (IUGR) are lower than those in control pregnant women, in the first, second [43,45] and third trimesters [48]. Women with in vitro fertlization (IVF) failure manifested by spontaneous abortion in the early pregnancy present lower sHLA-G in the pre-ovulation period and during pregnancy compared to women with normal pregnancies [49].

The reduction of HLA-G molecules could disregulate uterine natural killer (uNK) cells which are supposed to participate in the process of placentation and in uterine spiral artery transformation. Soluble HLA-G may contribute to trigger functional maturation of the uNK cells and vascular remodelling and decidualization. The reduced release of sHLA-G into the maternal circulation in preeclampsia and IUGR may alter the maternal-fetal immune relationship and thus be involved in the cause of these disorders.

Table 1. Published studies of HLA-G expression in serum of normal and pathological pregnant women.

Study

(year) Sample Technique MoAb Results

Normal condition Hunt JS

Am J Obstet Gynecol 2000

Serum

44 non pregnant women 129 pregnant women

ELISA 16G1, 16A1 sHLA-G higher in pregnant women

Hackmon R

Fetal Diagn Ther. 2004

Serum

21 pregnant women 16-20 weeks 19 women at term

ELISA 87G, 16G1 sHLA-G lower toward term

Yie SM

Am J Obstet Gynecol (2005)

Serum

12 pregnant women first, second, third trimesters ELISA 4H84, 3C/G4

sHLA-G decrease in third trimester pregnant women

Steinborn A

Am J Reprod Immunol (2007)

Plasma

40 non pregnant women 291 pregnant women

ELISA MEM-G9 sHLA-G increase in first trimester pregnant women

Rizzo R

Am J Reprod Immunol (2009)

Plasma

43 Pregnant women third trimester, at delivery ELISA MEM-G9 sHLA-G increase at delivery

Pathological condition Pfeiffer KA

Hum Immunol (2000)

Serum

65 IVF patients preovulatorily, after a positive HCG test weekly until the 9th gestational week

ELISA TP25.99 depletion W6/32

sHLA-G decrease in the pre-ovulation period and during pregnancy in women with a spontaneous abortion in IVF

Yie SM

Am J Obstet Gynecol (2005)

Serum

12 pregnant women and 12 PE women first, second, third trimesters

ELISA 4H84, 3C/G4 sHLA-G decrease in first trimester preeclamptic women

Steinborn A

Am J Reprod Immunol (2007)

Plasma

40 non pregnant women

291 pregnant women first, second, third trimesters 236 PE/IUGR pregant women first, second, third

trimesters

ELISA MEM-G9 sHLA-G decrease in second trimester preeclamptic women

Hackmon R

Am J Obstet Gynecol (2008)

Serum

24 pregnant women third trimester 26 PE pregnant women third trimester

ELISA MEM-G9 sHLA-G decrease in third trimester preeclamptic women

5.1. HLA-G and trophoblasts

The trophoblast differentiates into extravillous and villous tissues where extravillous cytotrophoblast (evct) cells invade the deciduas while villous cytotrophoblasts (vct) produce the outer villous syncytiotrophoblast (st) layer of chorionic villi (Figure 4). None of these villous trophoblast populations constitutively express HLA-A and -B at their surface. The absence of classical HLA class Ia molecules in cytotrophoblast cells, except for HLA-C, could activate NK cells towards fetal tissues. This is not the fact as the NK cells cytotoxicity is controlled by the interaction of C and HLA-G molecules with inhibitory receptors. The HLA-C1 group interacts with inhibitory KIR receptors 2DL2 and 2DL3, while the HLA-C2 group interacts with 2DL1 inhibitory receptor. HLA-G antigens interact with KIR2DL4 NK receptor [50] inducing proliferation and interferon (IFN)-γ secretion which might contribute to implantation and decidualization during early pregnancy [51].

HLA-G has been firstly detected in placenta by Ellis et al. [52] who have reported an HLA-G expression in chorionic membrane (extravillous) cytotrophoblast cells and in term amniochorion and trophoblast cells (Table 2). HLA-G has been observed in all types of extravillous cytotrophoblasts, with an increased gradient of expression from the villi into thedeciduas [53]. Non-trophoblastic HLA-G expression has been detected in Hofbauer cells in the mesenchymal core of chorionic villi [54] and in endothelial cells [55]. An RNAse protection assay and quantitative RT-PCR studies have indicate that the full-length transmembrane form of HLA-G (HLA-G1) is the predominant splice variant transcribed in vivo by trophoblasts and it is mainly expressed as a disulphide-linked homodimer [37]. Soluble HLA-G might derive from transmembrane HLA-G1 molecules and from HLA-G5 transcript [56-58]. Some investigators have failed to report HLA-G5 in villous placenta [59] underlining some differences in the results obtained by different research groups. It is of interest the work by Ishitani et al. [56] and Le Maoult et al. [58] which have verifiedthe presence of HLA-G5 isoform by blocking the staining with 16G1 monoclonal antibody (moAb) by theaddition of the 20-mer synthetic peptide and using a new

anti-HLA-G5/-G6moAb called 5A6G7 respectively. However these discrepancies are still to be resolved.

The importance of HLA-G presence in placental trophoblasts is evident in preeclampsia (Table 2), that is characterized by a defect in placental membrane and soluble HLA-G expression [60,61], where the absence of HLA-G molecules has an effect on fetal protection and vascular events.

Figure 4. HLA-G expression by cytotrophoblast cells in the early gestation. Placental villus, trophoblastic column, trophoblastic shell and deciduas are represented. HLA-G is expressed by invasive cytotrophoblast cells.

Table 2. Published studies of HLA-G expression in trophoblast cells of normal and pathological pregnant women.

Decidual vessel Invasive cytotrophoblasts (evct) Blastocyst Uterine epithelium Cytotrophoblast (vct) Syncitiotrophoblast (st) HLA-G Villus Colum Shell Interstitia Endo-vascular

Study

(year) Sample Technique MoAb Results

Normal condition Ellis SA

Immunology (1986) Amniochorion cytotrophoblast

Immunoflorescence SDS-PAGE Isoelectric focusing

W6/32 BBM.1

Novel HLA Class I molecule on chorionic cytotrophoblast cell membranes

Ishitani A

J Immunol (2003)

First trimester placenta(5–12 wk of gestation) and normal term placentas

(37–39wk of gestation) ELISA Immunohistochemical staining Western analysis 87G 01G 16G1

MembraneHLA-G in extravillous trophoblasts, soluble HLA-Gin all

placental trophoblasts

Morales PJ

J Immunol (2003) First trimester and term placentas

Immunohistochemical staining 1-2C3 (G1) 26-2H11 (G2)

sHLA-G1 and m/s-G2 are produced in placentas

m/sHLA-G2 present in the invasive trophoblast

Blaschitz A

Mol Hum Reprod (2005)

Term placentas

RT PCR

Immunohistochemistry and immunocytochemistry

ELISA Western blot analysis

MEM-G/1 4H84

Trophoblasts express only the HLA-G1 isoform

LeMaoult J

Mol Hum Reprod (2005)

First-trimester

extravillous cytotrophoblast Immunohistochemistry 5A6G7

HLA-G5 expressed by extravillous cytotrophoblasts

Apps R

Eur J Immunol (2007)

Decidual and placental tissues from first-trimester pregnancies

Immunoblotting Flow cytometry

G233 87G MEM-G/11

HLA-G - b2m-associated dimers expressed by trophoblasts

Pathological condition Goldman-Wohl D

Mol Hum Reprod (2000)

First, second trimesters and term placentas

12 uncomplicated pregnant women 10 PE pregnant women

RNA in-situ hybridization NA HLA-G mRNA expression defective in most preeclamptic placentae Term placentas

5.2. HLA-G and embryo

The first demonstration of HLA-G expression in preimplantation human embryos has been obtained by Jurisicova et al. [62] by RT-PCR and immunocytochemistry techniques (Table 3). The authors have shown the presence of a HLA-G heavy chain specific mRNA in about 40% of the 148 blastocystis tested. They have also reported that HLA-G mRNAis present in all preblastocyst development stages, including2- to 4-, 5- to 8-, and 9- to 16-cell embryos and morulas. The detection of HLA-G specific mRNA in preimplantation embryos has been confirmed by protein detection at cell membrane level and by an increased blastocystic cleavage rate when compared to embryos without HLA-G transcripts [63]. Yao et al. [64] have confirmed that human preimplantation embryosexpress HLA-G mRNA with some difficulties at the earliest stages but with an increasing proportion of positive embryos with developmental stage. The predominant isoform is HLA-G3 and -G4 whilethe full-length membrane bound (G1) and soluble forms (G5) and the truncated G2 and G6 vary in their expression,with G1 mRNA present in the 80% of blastocysts, soluble G5 in the 20% and soluble G6 in the 32%. Hunt et al. [65] have demonstrated that preimplantation human embryos express HLA-G5 but not HLA-G6 isoforms using specific HLA-G5 and –G6 moAbs [57]. These results are in striking contrast with mRNA data supporting the hypothesis of a gestational post-trascriptional programming of HLA-G isoforms. On the contrary Desoye et al. [66] have been unable to show a HLA-HLA-G staining on three unfixedpolyploid embryos at the 2-, 5-, and 8-cell stages; Roberts et al. [67] have not detected HLA-G onthree blastocysts and Hiby et al. [68] have found no HLA-G mRNA in 11preimplantation embryos ranging from the 2 cell to the blastocyststage using nested primers for full-length HLA-G. These contrasting data on HLA-G expression during the critical period of preimplantation embryonic development could be explained by differences in methodology and quality of the embryos used. Hiby et al. [68] have isolated mRNA from zona intact embryos with a standard phenol-chloroform extraction and a nested RT-PCR for full-length HLA-G with the outside forward primer located at exon 3 and the reverse primer at 3’ UTR,

primer sets cannot amplify HLA-G2 and -G3 isoforms because G2 lacks exon 3 and G3 lacks exon 3 and 4. On the contrary Yao et al. [64] have isolated mRNA with high performance magnetic beads removing the zona first. The quality of the embryos could be a possible reason for these different results. Yao et al. [64] have used only good-quality diploid embryos (grade A–C embryos and grade 4 blastocysts), which was not the case in previous studies [66].

The discrepancies obtained with embryo immunostaining for HLA-G molecules could be explained by the differences in methodology and the quality of the embryos used. Previous studies have used unfixed [66], acetone fixed [67], and paraformaldehyde fixed embryos [62,63] and a primary Ab incubation time between 30 min and 1 h. Yao et al. [64] and Shaikly et al. [69] have used the HLA-G-specific moAb MEMG/9 on fixed and permeabilized embryos, revealing both cytoplasmic and surface expression and prolonged overnight the Ab incubation to increase the sensitivity of the technique.

Shaikly et al. [69] have found a stronger HLA-G staining on the trophectoderm of blastocysts, reinforcing the hypothesis of an implication of HLA-G in early implantation of the embryo in the maternal uterus.

Table 3. HLA-G expression by preimplantation human embryos. Study (year) N. cultures N° positive

embryos (%) Culture media Hrs post fertilization Detection Antibody Fixative Positive results

Jurisicova A

PNAS (1996) ND ND Ham’s F10 24-96 1B8 Paraformaldehyde fixed

Yao YQ

J Immunol (2005) 20 15 (75) IVF 24-144 MEM-G9 Fixed and permeabilized

Shaikly VR

J Immunol (2008) 11 8 (73) Sequential medium 48-72 MEM-G9 Fixed and permeabilized

Negative results Desoye G

J Immunol (1988) 3 0 Ham’s F 10 24-96 W6/32 Unfixed

Roberts J

The first non-invasive proof of non classical HLA-I expression by human early embryos was obtained in 1999 by Menicucci et al. [70]. They have demonstrated the presence of soluble HLA-G molecules (sHLA-G) in 90% of 8-cell stage embryo culture supernatants, obtained by ART. A significant association has been observed between HLA-G production and embryo cleavage rate. This study was confirmed by Fuzzi et al. in 2002 [71] (Table 4). The authors have presented the first in vivo proof of the role of HLA-G molecules in pregnancy implantation showing the presence of soluble HLA-G molecules in the supernatants from cultures containing one to four embryos. Two groups of patients have been identified on the basis of sHLA-G molecule presence or absence in the embryo culture supernatants. Although no clinical differences have been observed between the two groups, positive embryo implantation occurred only in women with sHLA-G molecules in embryo culture supernatants.

This study started wide ranging research on this topic [72,73]. Further reports [69,74-86] have analyzed the presence of sHLA-G molecules in the supernatants from single embryo cultures by enzymatic immunosorbent assay (ELISA) and HLA-G specific monoclonal antibodies (MEM-G1; MEM-G9; 4H84, 3C/G4). They have obtained a significant relationship between the secretion of these molecules by early embryos and a higher implantation/pregnancy rate with a significantly higher proportion of sHLA-G positive embryos developing to blastocysts in vitro [69,82]. The meta-analysis of eleven studies evaluating sHLA-G in embryo culture for predicting pregnancy outcome in women undergoing ART has reported a modest diagnostic accuracy (DOR: 4.38 (95% CI, 2.93-6.55) while a subgroup analysis restricted to six studies with good quality embryos has shown an increase in the diagnostic performance (DOR: 12.67 (95% CI, 3.66-43.80)) [87]. These studies have investigated more than six thousand supernatants from single ART procedure embryos, with a recorded presence of sHLA-G from 36.2% [77] to 69.9% [80]. Shaikly et al. [69] have demonstrated a HLA-G specific labelling in the 50% of the twenty-four embryos analyzed with a distribution and intensity of fluorescence

Two main points should be focused on: (i) the presence of positive pregnancy outcomes also in sHLA-G negative samples. Sher et al. [75], Desai et al. [82] and Shaikly et al. [69] have reported significantly higher pregnancy rates when sHLA-G-positive embryos have been transferred but they have evidenced a pregnancy rate of 25-36% with sHLA-G-negative embryos. The absolute results of the first work by Fuzzi et al. [71], where all the positive pregnancies have been associated to sHLA-G positive embryo supernatants, could be ascribed to the presence of more embryos in each culture and/or to a different specificity of the detection assays. The different ELISA protocols used could account for different values and correlations of sHLA-G expression and implantation. Shaikly et al. [69] have confirmed the expression of sHLA-G in cleavage stage embryos during days 1 and 2 of human development but documented a marked variability of sHLA-G expression by early embryos from the same patient indicating a possible gestational embryo programming that could affect the results of sHLA-G and implantation correlation. However, sHLA-G-positive embryos have shown a higher rate of implantation in comparison with sHLA-G-negative embryos; (ii) the presence of sHLA-G is not indicative of chromosome normality [69,82] as no significant differences have been observed in sHLA-G expression between embryos diagnosed as chromosomally normal or abnormal. For this reason sHLA-G detection in conjunction with current morphological parameters to identify embryo implantation potential is needed.

In 2006 Ménézo et al. started a debate on the exact amount of sHLA-G produced by a single human embryo [88]. The authors have reported that human preimplantation embryo protein content is 45-50 ng with a consequent HLA-G release from 10 to 100% and above the total protein content of the embryo considering the levels of sHLA-G molecules in embryo supernatants claimed in the literature. The human embryo is able to produce high amounts of other proteins such as human chorionic gonadotropine (hCG) that can be detected in the mother’s urine [89] suggesting a massive production at the feto-maternal interface. A possible explanation of this protein production could be found in the morphological and metabolic changes of

(I) the cleavage stages are characterized by (i) a low metabolism, (ii) a high protein transduction rate and (iii) the necessity of exogenous pyruvate that are unusual features for any other mammalian somatic cell type [90];

(II) matrix metalloproteinases, the main proteinases facilitating the process of embryo implantation and uterus extracellular matrix remodeling and degradation [91], are present at all stages of embryo development from the one-cell to the blastocyst. It is known that metalloproteases enhance HLA-G shedding [92] suggesting the increased secretion of HLA-G by pre-implantation embryo as a result of metalloprotease activation;

(III) day-3 embryos cultured in vitro for 48 hours are able to secrete protein patterns similar to those of day-5 uterine blastocysts suggesting the in vitro culture is responsible for an accelerated embryo development and protein production [93].

These observations should be considered in the evaluation of the ability of embryonic cells to produce proteins as they have unique features and necessity in comparison with the other cell phenotypes. Their possibility to survive in the uterus is connected with their ability to escape the maternal immune system. Hence the importance, in the implantation process, to sustain an extensive HLA-G production. Two studies [94,95] have failed to detect sHLA-G molecules in embryo culture supernatants underlining some discrepancies with other studies (Table 4). The authors that have observed positive sHLA-G embryo cultures have documented different percentages of positive supernatants (20-70%) and levels from 38 pg/ml to 1890 ng/ml. Warner et al. [96] have reviewed the literature present in PubMed at the end of 2007. They found differences in the specificity and concentrations of capture and detection antibodies, in the characteristics of positive and negative controls, in the time and temperature of incubation. The embryo cultures, the time of collection and other clinical parameters were also compared. They have concluded that the presence of significant technical discrepancies could explain these contrasting results.

Some of the differences that should be taken into consideration when comparing these different studies are: (i) culture time; (ii) culture media; (iii) capture and detection antibodies; (iv) the standards (Table 4).

The embryo culture time is in a range from 48 to 120 hrs post fertilization. This could explain at least in part the different levels of sHLA-G. The 48 - 72 hour culture period seems to be the best time points to detect sHLA-G in embryo supernatants and was selected in the majority of the studies. Where different culture time were used contrasting results as soluble HLA-G are predictable as secretion levels measuredby ELISA may increase/decrease over time.

Culture media are also important for in vitro embryo growth [97] and can be divided accordingly to their composition in four groups: (i) glucose and inorganic phosphate-free medium (P1, IVC-One); (ii) low-glucose medium (IVF, Sequential medium); (iii) glucose, gentamicin sulphate and protein-free medium (Human Tubal Fluid (HTF)); (iv) early cleavage medium (ECM). The different composition of these culture media could have influenced sHLA-G production. Culture media play an important role in determining whether the embryo potentialcan be realized [98] leadingto epigenetic changes in the embryonic genome [99] and influencegene expression [98,100,101,102]. Rinaudo P and SchultzR [101] have observed that the expression of 114 genes, including genes involved in protein synthesis, cell proliferation and transporter functions, are affectedin embryoafter in vitro culture. Optimizing a culture medium in terms of its abilityto promote embryo growth [103] seems to avoid certain postnatal developmental and behavioural consequences and imply that minor variations in the culture media can lead to differences on the resulting embryos. For example human tubal fluid (HTF) and preimplantation stage one (P1) culture media differ in fertilization rate, embryo quality, implantation and pregnancy rates. Artini et al. [104] have demonstrated that embryo fertilization rate with HTF was 58.6% while with P1 62.5% (P = 0.003), the HTF embryo quality was lower (15.4%) than P1 embryos (68.7%) (P = 0.002) as the implantation rate (HTF embryos 6.8% versus P1 embryos 12.2%) (P = 0.02) and the pregnancy rate (HTF

different culture media could have influenced the results obtained analysing sHLA-G production by early embryos.

The discrepancy between the different ELISA systems may be related to a lack of specificity associated with cross-reactivity among capture and detection antibodies. Some of the MoAbs that have been used in these studies have presented cross-reactivities: 4H84 MoAb has a cross-reactivity with HLA-Ia [105]; 5A6B MoAb seems to recognize the denaturated HLA-G heavy chain and to have an affinity also for other proteins; BFL.1 MoAb has a doubtful HLA-G specificity as it fails to react with HLA-G transfected cell lines [106]

The positive standards could be subdivided into: cell culture purified molecules and recombinant proteins. These molecules could present a different structural conformation with a different antibody affinity that could affect MoAb recognition.

Study N. cultures N. Women N° positive cultures (%)

sHLA-G

range (ng/ml) Culture media

Hrs post fertilization Capture Antibody Detection Antibody Standard Positive results Fuzzi B

Eur J Immunol (2002) 285 101 231 (26) 1.4 - 11 IVF 72 MEM-G9 (G1, G5) W6/32 biotin 721.221G supernatant Roussev Fertil Steril (2003) >60 30 6 (20) ND P1 72 MEM-G9

(G1, G5) W6/32 biotin JEG3 supernatant

Sher G

Reprod Biomed (2004) 1245 201 101 (64) ND P1 72

MEM-G9

(G1, G5) W6/32 biotin JEG3 supernatant

Yie SM

Fertil Steril (2005) 386 137 270 (69.9) 10 - 1890 IVC One 72

4H84 (G1-G7) 3C/G4 Purified HLA-G from placenta Desai N Reprod Biomed (2006) 712 83 309 (43) 3 - 10 HTF 72 MEM-G9

(G1, G5) W6/32 biotin Exbio standard

Rebmann V

Human Immunol (2007) 588 313 117 (20) 0.038 – 5.628 IVF 48, 72, 96

MEM-G9

(G1, G5) β2m Purified sHLA-G

Fisch JD

Fertil Steril (2007) ~2083 209 ND ND ECM 72

MEM-G9

(G1, G5) W6/32 biotin

Human amniotic fluid

Rizzo R

J Reprod Immunol (2007) 50 38 26 (52) 1.2 - 13.1 IVF 72

MEM-G9 (G1, G5) β2m biotin 721.221G supernatant Lédée N Am J Reprod Immunol (2007) Day 2: 309 Day 3: 276 ND Day 2: 71 (23) Day 3: 188 (68) ND ND 48-72 ND ND ND Shaikly VR J Immunol (2008) 166 26 80 (50) 1.75 – 3.5 Sequential medium 48-72 MEM-G9

(G1, G5) W6/32 biotin Exbio standard

Negative results van Lierop MJ 15 ND 0 0 ND 48-120 G233 (G1, G5) 56B (G1, G4, 56B biotin BFL.1 biotin Recombinant

5.3. HLA-G and oocyte

A growing knowledge of human embryos obtained during through ART has not been paralleled by a similar knowledge of human oocytes, despite it being widely recognized that embryogenesis is deeply affected by oocyte quality. The main reason for this is probably that the selection for ART is applied to embryos in order to choose the best among them to be transferred in uterus. This widely practiced embryo selection has prevented the need to acquire skills in oocyte selection. In some countries, new laws and rules on ART need to move towards oocyte selection and to identify valid tools to recognize the best oocytes to be used for fertilization [107]. Currently oocyte selection is performed by using morphological parameters, without a clear association with a positive pregnancy outcome [108]. The development of non-invasive methods for oocyte selection could be an important step in all the ART laboratories.

The oocyte quality is associated with early embryonic survival, the establishmentand maintenance of pregnancy and fetal development. The quality and the developmental competence of an oocyte is acquired during the maturation process, during progressive differentiation throughout folliculogenesis. The ability of the oocyte to mature, be fertilized and to develop into a viable embryo starts with oocyte growth during the first steps of follicular development and goes on until the final oocyte capacitation. Ovarian cyclic activity induces some primordial follicles to grow, however, most of these follicles degenerate through atresia and in growing follicles, only a subset of oocytes are competent and able to support meiosis, fertilization and early embryo development to the blastocyst stage. Growing lines of evidence suggest that oocyte competence relies on the storage of messenger RNAs and proteins that will support early stages of embryo development, before full activation of embryonic genome. It is known that fertilizedoocyte transcription is silenced in the early stages of embryo development and 90% oocyte maternal mRNAs degrade in the 2 cell stage. The store of proteins in the fertilized oocyte issufficient to support embryo development to the 8 cell stageuntil the activation of the embryonic genome [109].

surrounding somatic cells suggests a central role of the oocyte in the success of folliculogenesis [110].

The follicle activation is a fundamental event for the oocyte maturation and it is controlled by a fine tuning of inhibitory and stimulatory soluble factors. The follicular fluid represents the essential and specific microenvironment for the regulation of the ovary function and oocyte maturation [111] and a possible relationship has been proposed between specific follicular fluid components (sFas-sFas ligand system, TNF-alpha, Nitric oxide, Hyaluronan, Gelatinases) and ART outcome [112].

Rizzo R et al. [85] have analyzed sHLA-G molecules in the follicular fluids (FFs) and have found a significant correlation between sHLA-G presence in FFs and in the culture supernatants of the corresponding fertilized oocytes (Figure 5). Lédée et al. [86] and Shaikly et al. [69] have confirmed the presence of sHLA-G molecules in FFs in 96% and 47% respectively but they have failed to identify a correlation with early embryo sHLA-G production (Table 5). Several differences in embryo culture conditions and the technical procedures could explain the differences in the correlation results. The presence of sHLA-G in FFs is not a confirmation that it is important in oocyte maturation but the presence of HLA-G molecules in follicular fluids suggests the possible role of this antigen in the oocyte maturation process. Further studies are required to confirm the relationship between FFs and embryo sHLA-G production.

Rizzo et al. [85] and Shaikly et al. [69] have identified granulosa cells as producers of HLA-G molecules (Figure 5), while there are conflicting results on HLA-G protein expression by oocytes. Desoye et al. [66] have found no staining on three unfixed oocytes, while Roberts et al. [67] have found 2 of 11 to be positive and Jurisicova et al. [62,63]have shown a positivestaining of 21 out of 33 oocytes. These findings have been confirmed by thepresence of HLA-G mRNA in 17 of 21 pools of 5-8 unfertilizedoocytes [62,63].

supernatants of 152 oocytes matured in vitro for sHLA-G presence. The in vitro oocyte maturation procedure has allowed the analysis of sHLA-G production by the cumulus-oocyte complex (COC) without the influence of the in vivo maternal microenvironment. The cumulus oophorus is characterized by the granulosa cells which surround the mammalian oocyte. These cells create a structural pathway for cell-to-cell communication where cumulus cells provide several trophic factors to the preovulatory oocyte [114]. Several results indicate that the measurement of gene transcription levels in cumulus cells would reliably complement the morphological oocyte evaluation providing a useful tool for selecting oocytes with greater chances to be fertilized [115,116].

Rizzo et al. [113] have demonstrated the the COCs produce sHLA-G molecules during oocyte maturation process. The main point is that no sHLA-G molecules have been detected in the COC culture supernatants corresponding to immature COCs while the highest sHLA-G production has been shown in good grade COCs. Some maturated COCs have failed to secrete sHLA-G, underlining that sHLA-G is only one of the factors implicated in this process. Further studies are necessary to confirm the possible role of G molecules in oocyte maturation and to evaluate if HLA-G could be a marker of oocyte quality.

Figure 5. HLA-G expression in mature follicle. HLA-G molecules are present in follicular fluid as evidenced by Western Blot analysis with MEM-G9 moAb (G1, G5: positive controls; FF1, FF2: follicular fluid; FO1, FO2: fertilized oocyte; HeLa: negative control) [85], in granulosa cells and polymorphonuclear cells as demonstrated by immunocytochemistry with MEM-G9 moAb [85] and in cumulus-oocyte complex as shown by Western Blot analysis with MEM-G9 moAb (+: positive control, -: negative control; O1, O2: cumulus-oocyte complex) [113].

Antrum Cumulus-oocyte complex Follicular Fluid Granulosa cells Polymorphonuclear-like cell Follicle From Rizzo R et al 2007 [85] From Rizzo R et al 2009 [113] HLA-G From Rizzo R et al 2007 [85]

Table 5. Soluble HLA-G expression in follicular fluids (FFs) and by the corresponding preimplantation human embryos. Study N. FFs N. Women N° positive embryo cultures (%) sHLA-G range in embryo (ng/ml) N° positive FFs (%) sHLA-G range in FFs (ng/ml) Culture media Hrs post fertilization Capture Antibody Detection Antibody Standard Rizzo R

J Reprod Immunol (2007) 50 38 26 (52) 1.2 - 13.1 19 (38) 12.8-162.0 IVF 72

MEM-G9 (G1, G5) β2m biotin 721.221G supernatant Lédée N Am J Reprod Immunol (2007) 117 ND Day 2: 20 (23) Day 3: 58 (68) ND 82 (96) ND ND 48 - 72 ND ND ND Shaikly J Immunol (2008) 60 12 13 (21) 1.75 – 3.5 29 (47) 1.75-4.0 IVF 72 MEM-G9 (G1, G5) W6/32

6. Expected HLA-G impact in ART

Currently, different approaches are used to select oocytes and embryos for ART procedures, but they do not assure a significant association with the pregnancy outcome.

Several studies have analyzed the possible implication of HLA-G molecules in the selection of embryos and oocytes for ART. Further analysis and novel approaches are necessary to overcome the contrasting results obtained by different researchers [117]. The possible use of HLA-G antigens as a marker for oocyte and embryo selection is auspicable as it could firstly increase the chances of success with ART, secondly allow the selection of the best oocyte/embryo to transfer so reducing the number of transferred embryos and hence the incidence of multiple pregnancies.

7. Expert commentary

Further research is needed to evaluate the potential of sHLA-G as a marker of oocyte/embryo competency and to provide definitive proof that soluble HLA-G molecules are secreted by early embryos. Clearly the analysis of HLA-G molecules has to be improved given the contrasting results obtained in trophoblasts cells [59]. Unfortunately, it is not currently possible tomeasure HLA-G mRNA and membrane-bound protein in the same oocyte/embryo,which would help to clarify these issues. The possible approach could be:

(i) a multicentre study where different laboratories could compare their results on the same samples.

Several technical workshops have been organizedin 2000, 2003 and 2004 to validate tools and protocols for HLA-Ganalysis [118,119] and laboratories are trying to agree standardization, in accordance with the Essen workshop in 2004 and the EMBIC Workshop on sHLA-G and Embryo Implantation in Oxford (June 2008).

(ii) the use of different techniques to confirm both positive and negative results. Several researchers are trying to develop new tools to analyze HLA-G molecules.

molecules; recently Rebmann et al. [83] have established a rapid detection assay based on Luminex technology, allowing sHLA-G quantification in sample volumes of only 10µl within 1.5 hours. These techniques could be an important improvement in order to increase the specificity and sensitivity of sHLA-G detection assays.

Further studies need to answer to the following questions: (i) What are the functions of sHLA-G during ooyte maturation and embryo implantation?, (ii) Why do some preimplantation embryos secrete sHLA-G and others not?, (iii) Where do sHLA-G molecules in FFs come from?

8. Five-year view

Much remains to be learned about the expression, regulation and functions of HLA-G gene products at the junction of fetal and maternal tissues. Cellular sources remain unclear and target cells are still not well defined.

How the semi-inflammatory conditions of the maternal–fetal interface might influence HLA class Ib gene expression is unknown. The functions of these antigens at the interface, during the embryo implantation and oocyte maturation remain to be precisely defined.

Despite these uncertainties, much progress has been made. It now seems likely that HLA-G is not simply an evolutionary remnant but is, instead, a major player in the establishment of an appropriate immunologic state for semiallogeneic pregnancy. Despite the considerable progress in HLA-G detection there are some problems in its analysis in oocyte/embryo culture supernatants.

Nowadays it is still mandatory to evaluate oocyte/embryo morphological parameters as research should confirm the role of HLA-G molecules in oocyte/embryo development.

The value of sHLA-G molecules as a marker of oocyte/embryo competency is extremely important as, in contrast with other techniques such as Pre-Implantation Genetic Diagnosis (PGD), this test is performed in a completely non-invasive way by

removing a small amount of the culture media surrounding two- and three-day-old embryos, and testing for sHLA-G.

The sHLA-G molecule is a research response to the need of a rational basis to select few and possibly single competent oocyte/embryo at a time, while maintaining optimal ART success rates.

Discovery of additional molecular markers [120] of oocyte/embryo competency and health will improve the potentiality of these non-invasive methodologies with wide possibilities for research and therapeutic application.

The future of ART will be to combine morphologic evaluations with a biochemical assessment of molecules that represent a marker of oocyte/embryo competency. The expected impact will be to increase the pregnancy rates that will be possible when the success rates will achieve a high quality value using these approaches and the detection of these molecules will be no more questionable. The ART laboratories could use morphologic parameters and biochemical markers for single embryo transfer, reducing the risk of multiple pregnancies.

However further research is needed to identify the real oocyte/embryo competence markers. HLA-G molecules could be one of them but it is mandatory to improve a standardized detection in order to obtain comparable results prior to use HLA-G as a oocyte/embryo selection marker.

Key issues

• Currently it is difficult to determine the more appropriate oocyte/embryo for transfer in ART protocols

• sHLA-G detection in oocyte/embryo culture supernatants could be a non-invasive method to determine oocyte/embryo quality

• Embryos with sHLA-G production are more likely to result in a successful implantation

• Possible correlation between sHLA-G detection in follicular fluids and in the corresponding fertilized oocyte

• sHLA-G detection in follicular fluids should be further analyzed to identify the possible role as marker for oocyte selection

• Possible correlation between sHLA-G detection in mature COCs

• sHLA-G detection in COC supernatants should be further analyzed to identify the possible role as marker for oocyte maturation

Acknowledgment

Thanks to Prof. Olavio Baricordi, Dr Marina Stignani, Dr Loredana Melchiorri and Dr. Amanda Neville for writing assistance.

References

1. Steptoe PC. Clinical aspects of pregnancies established with cleaving embryos grown in vitro. BR. J. Obstet. Gynecol. 87, 757-768 (1980).

2. Schultz RM. The science of ART. Science 296, 2188-2190 (2002).

3. Sutcliffe AG. IVF children: The first generation. Parthenon publishing: New York (2002).

4. Redman CW, McMichael AJ, Stirrat GM, Sunderland CA, Ting A. Class 1 major histocompatibility complex antigens on human extra-villous trophoblast. Immunology 52(3), 457-68 (1984).

5. Geraghty DE, Koller BH, Orr HT. A human major histocompatibility complex class I gene that encodes a protein with a shortened cytoplasmic segment. Proc Natl Acad Sci U S A 84(24), 9145-9149 (1987).

6. Warner CM, Panda P, Almquist CD et al. Preferential survival of mice expressing the Qa-2 antigen. J. Reprod. Fertil. 99(1), 145-147 (1993).

7. Comiskey M, Goldstein CY, De Fazio SR, Mammolenti M, Newmark JA, Warner CM. Evidence that HLA-G is the functional homolog of mouse Qa-2, the Ped gene product. Hum. Immunol. 64(11), 999-1004 (2003).

8. Rajagopalan S, Bryceson YT, Kuppusamy SP, et al. Activation of NK cells by an endocytosed receptor for soluble HLA-G. PLoS Biol. 4, e9 (2006).

9. Brown D, Trowsdale J, Allen R. The LILR family: modulators of innate and adaptive immune pathways in health and disease. Tissue Antigens 64, 215e25 (2004).

10.Carosella ED, Moreau P, Lemaoult J, Rouas-Freiss N. HLA-G: from biology to clinical benefits. Trends Immunol. 29(3), 125-132 (2008).

11.Hviid TV, Rizzo R, Melchiorri L, Stignani M, Baricordi OR. Polymorphism in the 5’ upstream regulatory and 3’ untranslated regions of the HLA-G gene in relation to soluble HLA-G and IL-10 expression. Hum. Immunol. 67, 53–62 (2006).

12. Hviid TV, Hylenius S, Rørbye C, Nielsen LG. HLA-G allelic variants are associated with differences in the HLA-G mRNA isoform profile and HLA-G mRNA levels. Immunogenetics 55, 63–79 (2003).

13.Rizzo R, Hviid TV, Stignani M, et al. The HLA-G genotype is associated with IL-10 levels in activated PBMCs. Immunogenetics 57(3-4), 172-181 (2005). 14.Hviid TV, Rizzo R, Christiansen OB, Melchiorri L, Lindhard A, Baricordi OR.

HLA-G and IL-10 in serum in relation to HLA-G genotype and polymorphisms. Immunogenetics 56, 135–141 (2004).

15.Chen XY, Yan WH, Lin A, Xu HH, Zhang JG, Wang XX. The 14 bp deletion polymorphisms in HLA-G gene play an important role in the expression of soluble HLA-G in plasma. Tissue Antigens 72, 335-341 (2008).

16.Rousseau P, Le Discorde M, Mouillot G, Marcou C, Carosella ED, Moreau P. The 14 bp deletion-insertion polymorphism in the 3' UT region of the HLA-G

gene influences HLA-G mRNA stability. Hum. Immunol. 64(11), 1005-1010 (2003).

17.Solier C, Mallet V, Lenfant F, Bertrand A, Huchenq A, Le Bouteiller P. HLA-G unique promoter region: functional implications. Immunogenetics 53, 617-625 (2001).

18.Lefebvre S, Berrih-Aknin S, Adrian F, et al. A specific interferon (IFN)-stimulated response element of the distal HLA-G promoter binds IFN-regulatory factor 1 and mediates enhancement of this nonclassical class I gene by IFN-beta. J. Biol. Chem. 276(9), 6133-6139 (2001).

19.Gobin SJ, Biesta P, de Steenwinkel JEM, Datema G, van den Elsen P. HLA-G transactivation by cAMP-response element-binding protein (CREB). J. Biol. Chemistry 277(42), 39525-31 (2002).

20.Moreau P, Mouillot G, Rousseau P, Marcou C, Dausset J, Carosella ED. HLA-G gene repression is reversed by demethylation. Proc Natl Acad Sci USA 100, 1191-1196 (2003).

21. Mouillot G, Marcou C, Rousseau P, Rouas-Freiss N, Carosella ED, Moreau P. HLA-G gene activation in tumor cells involves cisacting epigenetic changes. Int. J. Cancer 113, 928-936 (2005).

22.Hviid TVF, Møller C, Sørensen S, Morling N. Co-dominant Expression of the HLA-G Gene and Various Forms of Alternatively Spliced HLA-G mRNA in Human First Trimester Trophoblast. Human Immunol. 59(2), 87-98 (1998). 23.Copeman J, Han RN, Caniggia I, McMaster M, Fisher SJ, Cross JC.

Posttranscriptional regulation of human leukocyte antigen G during human extravillous cytotrophoblast differentiation. Biol. Reprod. 62(6), 1543-1550 (2000).

24.Carosella ED, Moreau P, Le Maoult J, Le Discorde M, Dausset J, Rouas-Freiss N. HLA-G molecules: from maternal-fetal tolerance to tissue acceptance. Adv. Immunol. 81, 199-252 (2003).

25.Dong Y, Lieskovska J, Kedrin D, Porcelli S, Mandelboim O, Bushkin Y. Soluble nonclassical HLA generated by the metalloproteinase pathway. Hum. Immunol. 64, 802-810 (2003).

26.Comiskey M, Domino KE, Warner CM. HLA-G is found in lipid rafts and can act as a signaling molecule. Hum. Immunol. 68(1), 1-11 (2007).

27. Lin A, Yan WH, Xu HH, et al. HLA-G expression in human ovarian carcinoma counteracts NK cell function. Ann. Oncol. 18(11), 1804-9 (2007). 28.Contini P, Ghio M, Poggi A, et al. Soluble HLA-A,-B,-C and -G molecules

induce apoptosis in T and NK CD8 cells and inhibit cytotoxic T cell activity through CD8 ligation. Eur. J. Immunol. 33, 125–34 (2003).

29. Lila N, Rouas-Freiss N, Dausset J, Carpentier A, Carosella ED. Soluble HLA-G protein secreted by allo-specific CD4 T cells suppresses the allo-proliferative response: a CD4 T cell regulatory mechanism. Proc. Natl. Acad. Sci. USA 98, 12150–12155 (2001).

30. Ristich V, Liang S, Zhang W, Wu J, Horuzsko A. Tolerization of dendritic cells by HLA-G. Eur. J. Immunol. 35, 1133–42 (2005).

31.Le Rond S, Azéma C, Krawice-Radanne I, et al. Evidence to support the role of HLA-G5 in allograft acceptance through induction of immunosuppressive/regulatory T cells. J. Immunol. 176, 3266–3276 (2006). 32.Fons P, Chabot S, Cartwright JE, et al. Soluble HLA-G1 inhibits angiogenesis

through an apoptotic pathway and by direct binding to CD160 receptor expressed by endothelial cells. Blood 108, 2608–2615 (2006).

33.Shiroishi M, Kuroki K, Rasubala L, et al. Structural basis for recognition of the nonclassical MHC molecule HLA-G by the leukocyte Ig-like receptor B2 (LILRB2/LIR2/ILT4/CD85d) Proc. Natl. Acad. Sci. USA. 103(44), 16412-16417 (2006).

34.Yan WH, Fan LA. Residues Met76 and Gln79 in HLA-G alpha1 domain involve in KIR2DL4 recognition. Cell Res. 15(3), 176-182 (2005).

35. Gonen-Gross T, Achdout H, Gazit R, et al., Complexes of HLA-G protein on the cell surface are important for leukocyte Ig-like receptor-1 function. J. Immunol. 171, 1343–1351 (2003).

36. Gonen-Gross T, Achdout H, Arnon TI, et al., The CD85J/leukocyte inhibitory receptor-1 distinguishes between conformed and beta 2-microglobulin-free HLA-G molecules. J. Immunol. 175, 4866–4874 (2005).

37. Apps R, Gardner L, Sharkey AM, Holmes N, Moffett A. A homodimeric complex of HLA-G on normal trophoblast cells modulates antigen-presenting cells via LILRB1. Eur. J. Immunol. 37(7), 1924-1937 (2007).

38. LeMaoult J, Caumartin J, Daouya M, et al. Immune regulation by pretenders: cell-to-cell transfers of HLA-G make effector T cells act as regulatory cells. Blood 109(5), 2040-2048 (2007).

39.Baricordi OR, Stignani M, Melchiorri L, Rizzo R. HLA-G and inflammatory diseases Inflamm. Allergy Drug Targets 7(2), 67-74 (2008).

40.Hunt JS. Stranger in a strange land. Immunol. Rev. 213, 36-47 (2006).

41.Medawar PB. Some immunological and endocrinological problems raised by the evolution of viviparity in vertebrates. Symp. Soc. Exp. Biol. 44, 320-338 (1953).

42.Piccinni MP. T cells in normal pregnancy and recurrent pregnancy loss. Reprod. Biomed. Online 13(6), 840-844 (2006).

43.Steinborn A, Varkonyi T, Scharf A, Bahlmann F, Klee A, Sohn C. Early detection of decreased soluble HLA-G levels in the maternal circulation predicts the occurrence of preeclampsia and intrauterine growth retardation during further course of pregnancy. Am. J. Reprod Immunol. 57(4), 277-286 (2007).

44.Hackmon R, Hallak M, Krup M, et al. HLA-G antigen and parturition: maternal serum, fetal serum and amniotic fluid levels during pregnancy. Fetal Diagn. Ther. 19(5), 404-409 (2004).